Tuning Real-World Image Restoration at Inference: A Test-Time Scaling Paradigm for Flow Matching Models

TL;DR

This paper introduces ResFlow-Tuner, a novel image restoration framework that combines flow matching models with test-time scaling and multi-modal fusion, achieving state-of-the-art results in real-world image restoration tasks.

Contribution

It proposes a training-free test-time scaling paradigm integrated with flow matching models for improved image restoration performance.

Findings

Achieves state-of-the-art results on multiple benchmarks.

Effectively leverages large pre-trained models for low-level vision.

Demonstrates significant performance gains with minimal computational overhead.

Abstract

Although diffusion-based real-world image restoration (Real-IR) has achieved remarkable progress, efficiently leveraging ultra-large-scale pre-trained text-to-image (T2I) models and fully exploiting their potential remain significant challenges. To address this issue, we propose ResFlow-Tuner, an image restoration framework based on the state-of-the-art flow matching model, FLUX.1-dev, which integrates unified multi-modal fusion (UMMF) with test-time scaling (TTS) to achieve unprecedented restoration performance. Our approach fully leverages the advantages of the Multi-Modal Diffusion Transformer (MM-DiT) architecture by encoding multi-modal conditions into a unified sequence that guides the synthesis of high-quality images. Furthermore, we introduce a training-free test-time scaling paradigm tailored for image restoration. During inference, this technique dynamically steers the denoising direction through feedback from a reward model (RM), thereby achieving significant performance gains with controllable computational overhead. Extensive experiments demonstrate that our method achieves state-of-the-art performance across multiple standard benchmarks. This work not only validates the powerful capabilities of the flow matching model in low-level vision tasks but, more importantly, proposes a novel and efficient inference-time scaling paradigm suitable for large pre-trained models.